Lithium and Halogenoderma

Halogenodermas are skin eruptions that result from ingestion of or exposure to halogen-containing drugs. The most characteristic manifestation is a vegetating plaque with peripheral pustules. We report 2 cases of eruptions classified as halogenoderma based on characteristic clinical and histological findings in patients receiving long-term lithium therapy with no history of halogen exposure.

A 48-year-old man who had been treated with lithium carbonate for 8 years for bipolar disorder presented with a 3-month history of several erythematous, verrucous plaques on his arms and chest (Figure 1). He was taking no other medications. A biopsy specimen of a lesion showed a digitated and papillated surface with prominent pseudocarcinomatous epithelial hyperplasia containing numerous intraepidermal microabscesses. The dermis contained a dense, mixed inflammatory infiltrate composed of neutrophils, plasma cells, eosinophils, histiocytes, and scattered multinucleated giant cells.

A 52-year-old woman with a history of bipolar disorder who had been treated with lithium carbonate for 6 years presented with a 6-week history of verrucous, crusted plaques on her face, with surrounding erythema (Figure 2). She was taking no other medications. The lesions initially presented as pustules that evolved over several weeks into vegetating plaques. A biopsy specimen of a lesion showed pseudoepitheliomatous hyperplasia with a dense mixed infiltrate in the dermis containing primarily neutrophils. Intraepidermal as well as some dermal abscesses were noted.

Special stains for organisms were negative in both cases, as were fungal, bacterial, and mycobacterial cultures.

Halogenodermas usually occur in association with iodide or bromide ingestion, but similar responses have been seen with the use of fluoride- and chloride-containing preparations. The diagnosis is based on a combination of clinical and histopathological findings and history, as there are no pathognomonic laboratory or pathological findings. The differential diagnosis includes deep mycoses, pemphigus vegetans, mycosis fungoides, syphilis, tuberculosis, anthrax, squamous cell carcinoma, Sweet syndrome, and factitial dermatitis. The exact pathogenesis of halogenoderma is unclear.

To explore a possible connection between lithium ingestion and the development of halogenoderma, we reviewed the literature on lithium and its effects on thyroid gland function and serum iodide levels. The association between long-term lithium use and the development of goiter and, occasionally, hypothyroidism has been well documented.¹ It appears that lithium acts at the level of the thyroid gland to block release of thyroxine. The effect of lithium on iodide uptake into the thyroid gland has also been examined. Numerous studies have demonstrated an inhibitory effect of lithium on iodide uptake into the thyroid gland. In vitro studies using rat and porcine thyroid follicles have shown that lithium at a 2-mmol/L concentration suppresses thyrotropin-induced iodide uptake.² Similarly, an inhibitory effect on iodide uptake as measured by a reduced thyroid-plasma ratio of iodine 125 in lithium-treated quails as opposed to controls has been reported.³ Iodide uptake into the thyroid occurs via an active transport mechanism closely related to the function of the sodium-potassium–adenosine triphosphatase system. Thyrotropin-mediated stimulation of adenylate cyclase with subsequent generation of cyclic adenosine monophosphate is the most important factor enhancing iodide transport. The inhibition of iodide uptake by lithium may be related to its inhibitory effect on adenylate cyclase, the same mechanism believed to be primarily responsible for its adverse effect on thyroxine release.⁴ Alternatively, lithium may decrease iodide uptake via inhibition of the sodium-potassium pump.⁵ Inhibition of iodide uptake may result in increased serum levels of iodide, potentially explaining the development of halogenoderma in our patients.

As this hypothesis was generated in retrospect, measurement of serum iodide levels, a critical step in testing its validity, was not performed because the patients were no longer available for follow-up. We report our findings to make others aware of a possible rare association between lithium and the development of halogenoderma and to suggest that laboratory investigations, including thyroid function tests and serum iodide measurement, may be appropriate in such cases.